Initial ITA timeline and whole trace
In your RoundRobin directory, typing ls will show you a file named roundRobin.stf.
Exercise: Do this!
[agopu@bc81 agopu]$ cd ~/MPI_Tutorial/RoundRobin/ [agopu@bc81 RoundRobin]$ ls -l roundRobin.s* -rw------- 1 agopu hpc 10117 Feb 7 12:03 roundRobin.stf -rw------- 1 agopu hpc 1335 Feb 7 12:03 roundRobin.stf.dcl -rw------- 1 agopu hpc 32728 Feb 7 12:03 roundRobin.stf.frm . . . -rw------- 1 agopu hpc 32728 Feb 7 12:03 roundRobin.stf.sts
These trace files were created automatically by Intel Trace Collector when the roundRobin program was executed as part of an exercise in the RoundRobin - Simple Message Passing Illustrated section.
No modification of the source code roundRobin.c was required. The linker flag -lVT to the libVT.a Intel Trace Collector library in the Makefile did the needful.
An important tool that enables the parallel programmer to view the lines of communication among processors is the trace display tool Intel Trace Analyzer (ITA), formerly called Vampir Trace. ITA can be used to examine roundRobin.stf.
Exercise: Do this!
At the shell prompt (on interactive nodes you got through qsub -I ), type:
[agopu@bc81 RoundRobin]$ traceanalyzer roundRobin.stf &
You should first see an initial (rough) ITA activity timeline for the roundRobin program run on two nodes (four processes):
The basic elements of the initial timeline are:
The horizontal axis is time.
Each horizontal bar corresponds to a process or a set of processes. In our case, since we were using two nodes and two processors in each of those nodes, ITA merges the timelines on a per node basis. At the left margin they are clearly listed - notice how processes 0 and 2, which ran on node bc43, are represented on the second bar. Soon, we'll show how it's possible to look at each process in finer detail.
Red segments show execution of MPI functions. Green segments show execution of the routines of the program. Neither the overhead imposed by the ITC/A tracing application itself nor the messages exchanged between processors is not shown in this initial timeline.
- Whole Trace :
Exercise: Do this!
Next, right-click on the initial timeline frame and select Load -> Whole Trace . This will open up another summary chart - you can see how much time MPI calls took and so forth for the entire application. Shown below is an example of what you could expect to see for the roundRobin program:
